Approving a Tool Is Not Approving Data: What Laravel AI SDK 1.0 Changes, and What It Doesn't

Automated excerpt

A tool declares Approvable, uses the InteractsWithApprovals trait, and when the model decides to call it the conversation stops and hands back the pending calls with the arguments the model picked: You resume by passing a decision per call: approve, reject with a reason the model gets to read, or edit the arguments before the tool runs. The mental model it suggests, though, is: the agent proposes, the human confirms, the action runs. Hence the rule I ended up with: human approval inside the agent protects whoever is driving the agent.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.